{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "using Compat, Random, Distributions, Plots, LinearAlgebra, CSV, DataFrames, SpecialFunctions, Statistics, NLsolve, Optim, KernelDensity, LaTeXStrings\n",
    "gr(fmt=:png)\n",
    "Random.seed!(1);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gap (generic function with 1 method)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "function norm_den(x)\n",
    "    return 1/sqrt(2*π*s^2)*exp(-(x-m)^2/2/s^2)\n",
    "end\n",
    "\n",
    "function gap(x, g)\n",
    "    F = zeros(4)\n",
    "    a1 = abs(x[1])\n",
    "    a2 = abs(x[2])\n",
    "    b = abs(x[3])\n",
    "    w = min(abs(x[4]),1)\n",
    "    F[1] = g[1] - (w*a1*(a1+1) + b^2*(1-w)*(a2+1)*a2)/(w*a1 + b*(1-w)*a2)^2\n",
    "    F[2] = g[2] - (w*a1*(a1+1)*(a1+2)+b^3*(1-w)*(a2+2)*(a2+1)*a2)/(w*a1*(a1+1) + b^2*(1-w)*(a2+1)*a2)/(w*a1 + b*(1-w)*a2)\n",
    "    F[3] = g[3] - (w*a1*(a1+1)*(a1+2)*(a1+3)+b^4*(1-w)*(a2+3)*(a2+2)*(a2+1)*a2)/(w*a1*(a1+1) + b^2*(1-w)*(a2+1)*a2)^2\n",
    "    F[4] = g[4] - (w*a1*(a1+1)*(a1+2)*(a1+3)*(a1+4) + b^5*(1-w)*(a2+4)*(a2+3)*(a2+2)*(a2+1)*a2)/(w*a1*(a1+1) + b^2*(1-w)*(a2+1)*a2)/(w*a1*(a1+1)*(a1+2) + b^3*(1-w)*(a2+2)*(a2+1)*a2)\n",
    "    return sum(F.*F)\n",
    "end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read the file\n",
    "\n",
    "df = CSV.read(\"usa1.csv\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fill the missing\n",
    "\n",
    "count, L = 0, size(df,1)\n",
    "new_prices = zeros(L)\n",
    "for i in df.id\n",
    "    count = count + 1\n",
    "    if count < L\n",
    "        if isa(df.miss[count+1],Number) & !isa(df.miss[count],Number) & (df.id[count] == df.id[count+1])\n",
    "            j = count + 1\n",
    "            while (df.id[j] == df.id[count]) & isa.(df.miss[j],Number)\n",
    "                j = j + 1\n",
    "            end\n",
    "            if (df.id[j] == df.id[count])\n",
    "                for k in count:1:j\n",
    "                    if k < (count+j)/2+1\n",
    "                        new_prices[k] = df.price[count]\n",
    "                        else\n",
    "                            new_prices[k] = df.price[j]\n",
    "                        end\n",
    "                end     \n",
    "                else\n",
    "                    for k in count:1:j-1\n",
    "                        new_prices[k] = df.price[count]\n",
    "                    end\n",
    "                end        \n",
    "        end\n",
    "    end\n",
    "end\n",
    "new_prices[isa.(df.price,Number)] = df.price[isa.(df.price,Number)];"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [],
   "source": [
    "#initiate the spells database\n",
    "\n",
    "spell_price = []\n",
    "spell_url = []\n",
    "spell_duration = []\n",
    "spell_id = []\n",
    "spell_category = []\n",
    "spell_sale = []\n",
    "spell_initial = []\n",
    "spell_final = []\n",
    "duration = 0\n",
    "init = 0\n",
    "fin = 0\n",
    "\n",
    "for i in 1:1:L-1\n",
    "    duration = duration + 1\n",
    "    if isa(df.initialspell[i],Number)\n",
    "        init = 1\n",
    "    end\n",
    "    if isa(df.lastspell[i],Number)\n",
    "        fin = 1\n",
    "    end\n",
    "    if (df.id[i+1] != df.id[i]) || (new_prices[i+1] != new_prices[i])\n",
    "        append!(spell_price,new_prices[i])\n",
    "        append!(spell_id,df.id[i])\n",
    "        append!(spell_url,df.cat_url[i])\n",
    "        append!(spell_duration,duration)\n",
    "        append!(spell_category,df.category[i])\n",
    "        append!(spell_initial,init)\n",
    "        append!(spell_final,fin)\n",
    "        if isa(df.sale[i],Number)\n",
    "            append!(spell_sale,df.sale[i])\n",
    "        else\n",
    "            append!(spell_sale,0)                \n",
    "        end\n",
    "        duration = 0\n",
    "        init = 0\n",
    "        fin = 0\n",
    "    end\n",
    "end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sales deleted: 0\n",
      "Short spells deleted: 0\n",
      "Small changes deleted: 0\n",
      "Large changes deleted: 87"
     ]
    }
   ],
   "source": [
    "# write the new dataframe\n",
    "\n",
    "dfspells = DataFrame(id = spell_id, url = spell_url, category = spell_category, price = spell_price, duration = spell_duration, initial = spell_initial, final = spell_final, sale = spell_sale)\n",
    "dfspells.change = zeros(size(dfspells,1))\n",
    "dfspells.logchange = zeros(size(dfspells,1))\n",
    "upper_cutoff = 1.5\n",
    "lower_cutoff = 0.009\n",
    "remove_sales, remove_short = false, false\n",
    "\n",
    "# replace the sales, small and large deviations, and short spells with the previous price\n",
    "\n",
    "flag = [0, 0, 0, 0, 0]\n",
    "N = length(dfspells.id)\n",
    "for i in 1:1:N-1\n",
    "    if (remove_sales) & (dfspells.sale[i+1] == 1)\n",
    "        dfspells.price[i+1] = dfspells.price[i]\n",
    "        flag[1] += 1\n",
    "    end\n",
    "    if (remove_short) & (dfspells.id[i] == dfspells.id[i+1]) & (dfspells.duration[i+1] < 3)\n",
    "        dfspells.price[i+1] = dfspells.price[i]\n",
    "        flag[2] += 1\n",
    "    end\n",
    "    if (abs(dfspells.price[i+1] - dfspells.price[i]) < lower_cutoff) & (dfspells.id[i] == dfspells.id[i+1])\n",
    "        if abs(dfspells.price[i+1] - dfspells.price[i]) > 0\n",
    "            flag[3] += 1\n",
    "        end\n",
    "        dfspells.price[i+1] = dfspells.price[i]\n",
    "    end\n",
    "    if (dfspells.id[i] == dfspells.id[i+1]) & (abs(log(dfspells.price[i+1]) - log(dfspells.price[i])) > upper_cutoff)\n",
    "        dfspells.price[i+1] = dfspells.price[i]\n",
    "        flag[4] += 1\n",
    "    end\n",
    "end\n",
    "print(\"Sales deleted: \",flag[1],\"\\nShort spells deleted: \",flag[2],\"\\nSmall changes deleted: \",flag[3],\"\\nLarge changes deleted: \",flag[4])\n",
    "\n",
    "# merge neighboring spells with the same price and id\n",
    "\n",
    "for i in 1:1:N-1\n",
    "    if (abs(dfspells.price[i] - dfspells.price[i+1]) < lower_cutoff) & (dfspells.id[i] == dfspells.id[i+1])\n",
    "        dfspells.duration[i+1] = dfspells.duration[i+1] + dfspells.duration[i]\n",
    "        dfspells.initial[i+1] = dfspells.initial[i]\n",
    "        dfspells.duration[i] = 0\n",
    "    end\n",
    "end\n",
    "\n",
    "# delete zero durations and add changes\n",
    "\n",
    "dfspells = dfspells[dfspells.duration .> 0, :]\n",
    "dfspells = dfspells[dfspells.initial.*dfspells.final .== 0, :]\n",
    "N = length(dfspells.id)\n",
    "for i in 1:1:N-1\n",
    "    if dfspells.initial[i+1] == 0\n",
    "        dfspells.change[i+1] = (dfspells.price[i+1]-dfspells.price[i])/dfspells.price[i]\n",
    "        dfspells.logchange[i+1] = log(dfspells.price[i+1])-log(dfspells.price[i])\n",
    "    end\n",
    "end\n",
    "select!(dfspells, Not(:sale))\n",
    "dfspells.length = dfspells.initial - dfspells.final\n",
    "\n",
    "# go forwards and write the spell number\n",
    "\n",
    "count = 0\n",
    "for i in 1:1:N\n",
    "    count = count + 1\n",
    "    dfspells.initial[i] = count\n",
    "    if dfspells.final[i] == 1\n",
    "        count = 0\n",
    "    end\n",
    "end\n",
    "\n",
    "# go backwards and write the spell length\n",
    "\n",
    "num = 0\n",
    "for i in 1:1:N\n",
    "    if dfspells.final[N+1-i] == 1\n",
    "        num = dfspells.initial[N+1-i]\n",
    "    end\n",
    "    dfspells.length[N+1-i] = num\n",
    "end\n",
    "\n",
    "# three products have confcting category identificators, delete them\n",
    "\n",
    "dfspells = dfspells[dfspells.id .!= 41976, :]\n",
    "dfspells = dfspells[dfspells.id .!= 46568, :]\n",
    "dfspells = dfspells[dfspells.id .!= 45549, :];\n",
    "\n",
    "# write the file with spells\n",
    "\n",
    "CSV.write(\"spells1.csv\",dfspells);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {},
   "outputs": [],
   "source": [
    "# take products with at least 3 spells (equivalently, at least two changes)\n",
    "\n",
    "df2 = dfspells[dfspells.length .> 2, :]\n",
    "S = size(df2[df2.initial .== 3, :],1)\n",
    "pow41 = (df2[df2.initial .== 2, :].logchange).^4\n",
    "pow21 = (df2[df2.initial .== 2, :].logchange).^2\n",
    "pow22 = (df2[df2.initial .== 3, :].logchange).^2;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 285,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# histogram for an exemplary category\n",
    "\n",
    "df561 = dfspells[dfspells.category .== 561, :]\n",
    "density = kde(df561[df561.logchange .!= 0.0, :].logchange,bandwidth=0.03,kernel=Logistic)\n",
    "histogram(df561[df561.logchange .!= 0.0, :].logchange,xlim=[-1.0,1.0],bins=200,legend=false,normed=true)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"561.csv\""
      ]
     },
     "execution_count": 286,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CSV.write(\"561.csv\",df561[df561.logchange .!= 0.0, :])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean absolute price change 0.22689574071501514;\n",
      " standard deviation of absolute price change 0.14301799817841288"
     ]
    },
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 287,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# symmetrized histogram and fitted kernel density\n",
    "\n",
    "changes = [df561[df561.logchange .!= 0.0, :].logchange; -df561[df561.logchange .!= 0.0, :].logchange]\n",
    "density = kde(changes,bandwidth=0.03,kernel=Logistic)\n",
    "kernel_den = DataFrame(x = density.x, q = density.density)\n",
    "CSV.write(\"kernel_den.csv\",kernel_den)\n",
    "print(\"mean absolute price change \",mean(abs.(changes)),\";\\n standard deviation of absolute price change \",std(abs.(changes)))\n",
    "histogram(changes,xlim=[-1.0,1.0],normed=true,bins=200,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 288,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "histogram(df561[df561.logchange .!= 0.0, :].logchange,xlim=[-1.0,1.0],bins=200,legend=false,normed=true)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot q(0) as a function of bandwidth\n",
    "\n",
    "BW, q0 = collect(0.001:0.0001:0.03), []\n",
    "for bw in BW\n",
    "    dens = kde(changes,bandwidth=bw)\n",
    "    append!(q0,(dens.density[Int64(length(dens.density)/2)+1]+dens.density[Int64(length(dens.density)/2)])/2)\n",
    "end\n",
    "plot(BW,q0,label=L\"q(0)\",legend=:topleft,color=:red,xlabel=\"bandwidth\")"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "# bootstrapping procedure to see the difference between the pooled kurtosis \n",
    "# and the underlying kusrtosis estimated with the first two price changes\n",
    "\n",
    "b, B, K, K_pooled = 10000, 5000, [], []\n",
    "p4, p21, p22 = zeros(b), zeros(b), zeros(b)\n",
    "\n",
    "for j in 1:1:B\n",
    "    ind = rand(DiscreteUniform(1,S),b)\n",
    "    count = 0\n",
    "    for i in ind\n",
    "        count += 1\n",
    "        p21[count] = pow21[i]\n",
    "        p22[count] = pow22[i]\n",
    "        p4[count] = pow41[i]\n",
    "    end\n",
    "    append!(K, mean(p4)/mean(p22.*p21))\n",
    "    append!(K_pooled, mean(p4)/mean(p21)^2)\n",
    "end\n",
    "\n",
    "m, s = mean(K), std(K)\n",
    "norm1 = norm_den.(sort!(collect(Set(K))))\n",
    "print(\"\\nUnderlying: \", m, \" (\", s, \")\\n\")\n",
    "m, s = mean(K_pooled), std(K_pooled)\n",
    "print(\"Pooled: \", m, \" (\", s, \")\\n\")\n",
    "norm2 = norm_den.(sort!(collect(Set(K_pooled))))\n",
    "\n",
    "histogram(K,label=\"Underlying\",legend=:topright,color=:blue,normalize=true)\n",
    "histogram!(K_pooled,label=\"Pooled\",color=:red,normalize=true)\n",
    "plot!(sort!(collect(Set(K))),norm1,linewidth=3,color=:black,label=\"\")\n",
    "plot!(sort!(collect(Set(K_pooled))),norm2,linewidth=3,color=:black,label=\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {},
   "outputs": [],
   "source": [
    "# initiate the category database\n",
    "\n",
    "cat_ids, number = collect(Set(df2.category)), []\n",
    "\n",
    "for cat in cat_ids\n",
    "    append!(number, length(collect(Set(df2[df2.category .== cat, :].id))))\n",
    "end\n",
    "\n",
    "statistics = DataFrame(category_id = Int64.(cat_ids), number_of_products = Int64.(number))\n",
    "statistics = statistics[statistics.number_of_products .> 100, :]\n",
    "statistics.number_price_changes = zeros(Int64,size(statistics,1))\n",
    "statistics.mean_price_change = zeros(size(statistics,1))\n",
    "statistics.mean_abs_change = zeros(size(statistics,1))\n",
    "statistics.mean_inv_change = zeros(size(statistics,1))\n",
    "statistics.std_price_change = zeros(size(statistics,1))\n",
    "statistics.skewness = zeros(size(statistics,1))\n",
    "statistics.alpha_with22 = zeros(size(statistics,1))\n",
    "statistics.alpha_with22_ste = zeros(size(statistics,1))\n",
    "statistics.alpha_with31 = zeros(size(statistics,1))\n",
    "statistics.alpha_with31_ste = zeros(size(statistics,1))\n",
    "statistics.g11 = zeros(size(statistics,1))\n",
    "statistics.g11_ste = zeros(size(statistics,1))\n",
    "statistics.g21 = zeros(size(statistics,1))\n",
    "statistics.g21_ste = zeros(size(statistics,1))\n",
    "statistics.g22 = zeros(size(statistics,1))\n",
    "statistics.g22_ste = zeros(size(statistics,1))\n",
    "statistics.g31 = zeros(size(statistics,1))\n",
    "statistics.g31_ste = zeros(size(statistics,1))\n",
    "statistics.g32 = zeros(size(statistics,1))\n",
    "statistics.g32_ste = zeros(size(statistics,1))\n",
    "statistics.M1 = zeros(size(statistics,1))\n",
    "statistics.M2 = zeros(size(statistics,1))\n",
    "statistics.M3 = zeros(size(statistics,1))\n",
    "statistics.M4 = zeros(size(statistics,1))\n",
    "statistics.underlying12_kurt = zeros(size(statistics,1))\n",
    "statistics.underlying12_kurt_ste = zeros(size(statistics,1))\n",
    "statistics.pooled_kurt = zeros(size(statistics,1))\n",
    "statistics.pooled_kurt_ste = zeros(size(statistics,1))\n",
    "statistics.corr12 = zeros(size(statistics,1))\n",
    "statistics.coefvar1 = zeros(size(statistics,1))\n",
    "statistics.coefvar2 = zeros(size(statistics,1))\n",
    "statistics.coefvar = zeros(size(statistics,1))\n",
    "statistics.implied_corr = zeros(size(statistics,1))\n",
    "statistics.mean_duration = zeros(size(statistics,1))\n",
    "statistics.ste_duration = zeros(size(statistics,1))\n",
    "statistics.alpha1 = zeros(size(statistics,1))\n",
    "statistics.alpha2 = zeros(size(statistics,1))\n",
    "statistics.ratio_betas = zeros(size(statistics,1))\n",
    "statistics.weight_on1 = zeros(size(statistics,1))\n",
    "statistics.C_pooled = zeros(size(statistics,1))\n",
    "statistics.C = zeros(size(statistics,1));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.350257671746138 -0.000789120115529093 0.000789120115529093\n",
      " category number  111 of the size 3439\n",
      "0.46571865065306406 -0.0007626789768787855 0.0007626789768787855\n",
      " category number  119 of the size 3228\n",
      "0.465455410917068 -0.0007670500941378205 0.0007670500941378205\n",
      " category number 1212 of the size 2551\n",
      "0.4802031806364933 -0.000699632796845384 0.000699632796845384\n",
      " category number  122 of the size 1405\n",
      "0.33922015047905807 -0.0007402449833504715 0.0007402449833504715\n",
      " category number  118 of the size 1390\n",
      "0.44337668400347074 -0.0007434879485456473 0.0007434879485456473\n",
      " category number  117 of the size 1154\n",
      "0.3389061985872859 -0.0007354522145824406 0.0007354522145824406\n",
      " category number  561 of the size 1032\n"
     ]
    }
   ],
   "source": [
    "# estimate the statistics for the seven biggest categories\n",
    "\n",
    "statistics = statistics[statistics.number_of_products .> 1000, :]\n",
    "sort!(statistics,:number_of_products,rev=true)\n",
    "cat, gammas = 1, zeros(size(statistics,1),5)\n",
    "\n",
    "for c in statistics.category_id\n",
    "    dfcat = df2[df2.category .== c, :]\n",
    "    \n",
    "    # use the first two price changes to estimate the underlying kurtosis and correlations\n",
    "    # bootstrap standard errors\n",
    "    S = size(dfcat[dfcat.initial .== 3, :],1)\n",
    "    pow41 = (dfcat[dfcat.initial .== 2, :].logchange).^4\n",
    "    pow42 = (dfcat[dfcat.initial .== 3, :].logchange).^4\n",
    "    pow21 = (dfcat[dfcat.initial .== 2, :].logchange).^2\n",
    "    pow22 = (dfcat[dfcat.initial .== 3, :].logchange).^2\n",
    "    \n",
    "    B, K, K_pooled, p41, p42, p21, p22  = 5000, [], [], zeros(S), zeros(S), zeros(S), zeros(S)\n",
    "    for j in 1:1:B\n",
    "        ind = rand(DiscreteUniform(1,S),S)\n",
    "        count = 0\n",
    "        p21, p22, p41, p42 = pow21[ind], pow22[ind], pow41[ind], pow42[ind]\n",
    "        append!(K, 0.5*mean(p41)/mean(p22.*p21) + 0.5*mean(p42)/mean(p22.*p21))\n",
    "        append!(K_pooled, mean(vcat(p41,p42))/mean(vcat(p21,p22))^2)\n",
    "    end\n",
    "    \n",
    "    # write the first set of statistics\n",
    "    i = findfirst(isequal(c),statistics.category_id)\n",
    "    statistics.number_price_changes[i] = length(dfcat.logchange)\n",
    "    statistics.mean_price_change[i] = mean(dfcat.logchange)\n",
    "    statistics.mean_abs_change[i] = mean(abs.(dfcat.logchange))\n",
    "    statistics.mean_inv_change[i] = mean(abs.(dfcat[dfcat.logchange.>0,:].logchange).^(-1))\n",
    "    statistics.std_price_change[i] = std(dfcat.logchange)\n",
    "    statistics.skewness[i] = mean((dfcat.logchange-mean(dfcat.logchange)*ones(length(dfcat.logchange))).^3)/(statistics.std_price_change[i])^3\n",
    "    statistics.mean_duration[i], statistics.ste_duration[i] = mean(dfcat.duration), std(dfcat.duration)\n",
    "    statistics.underlying12_kurt[i], statistics.underlying12_kurt_ste[i] = (0.5*mean(pow41)+0.5*mean(pow42))/mean(pow22.*pow21), std(K)\n",
    "    statistics.corr12[i], statistics.coefvar1[i], statistics.coefvar2[i] = cor(pow21,pow22), std(pow21)/mean(pow21), std(pow22)/mean(pow22)\n",
    "    \n",
    "    changes = [dfcat.logchange[abs.(dfcat.logchange).>0.0001]; -dfcat.logchange[abs.(dfcat.logchange).>0.0001]]\n",
    "    dens = kde(changes)\n",
    "    print(dens.density[Int64(length(dens.density)/2)],\" \",dens.x[Int64(length(dens.density)/2)],\" \",dens.x[Int64(length(dens.density)/2)+1],\"\\n\")\n",
    "    statistics.C_pooled[i] = dens.density[Int64(length(dens.density)/2)]/2*var(changes)/mean(abs.(changes))\n",
    "    \n",
    "    gamma11, gamma21, gamma22, gamma31, gamma32 = 0, 0, 0, 0, 0\n",
    "    Sigma31, Sigma22,  = zeros(3,3), zeros(3,3)\n",
    "    M1, M2, M3, M4, alpha = 0, 0, 0, 0, 0\n",
    "\n",
    "    # estimate the dimensionless statisticsistics using the algorithm\n",
    "    ids = collect(Set(dfcat.id))\n",
    "    cols = vcat(Symbol(\"product\"),[Symbol(\"e$k\") for k in [11, 21, 31, 32, 22]])\n",
    "    cols = vcat(cols,[Symbol(\"e$k\") for k in [2, 3, 4, 5]])\n",
    "    dfprod = DataFrame(vcat(Int64, [Float64 for k in 1:9]), cols, length(ids))\n",
    "    dfprod.product = ids\n",
    "    num_correction = []\n",
    "    for prod in ids\n",
    "        changes = abs.(dfcat[dfcat.id .== prod, :].logchange[2:end])\n",
    "        changes = changes[changes.>0]\n",
    "        p = findfirst(isequal(prod),dfprod.product)\n",
    "        l = length(changes)\n",
    "        \n",
    "        # correction for C\n",
    "        numerator, Z = mean(changes), (changes.^(-1))*transpose(changes.^2)\n",
    "        append!(num_correction,(sum(Z) - numerator*l)/l/(l-1))\n",
    "        \n",
    "        # gammas\n",
    "        numerator, summa, k, j = 0, 0, 0, 0\n",
    "        row = []\n",
    "        for pair in [11, 21, 31, 32, 22]\n",
    "            j = Int64(pair%10)\n",
    "            k = Int64((pair - j)/10)\n",
    "            numerator, Z = mean(changes.^(j+k)), (changes.^j)*transpose(changes.^k)\n",
    "            append!(row,(sum(Z) - numerator*l)/l/(l-1))\n",
    "            append!(row,numerator)\n",
    "        end\n",
    "        for col in 1:1:4\n",
    "            dfprod[p,1+col], dfprod[p,6+col] = row[2*col-1], row[2*col]\n",
    "        end\n",
    "        dfprod[p,6] = row[end-1]\n",
    "    end\n",
    "    \n",
    "    # bootstrap standard errors\n",
    "    mom1, mom2, mom3, mom4 = [], [], [], []\n",
    "    G11, G21, G22, G31, G32 = [], [], [], [], []\n",
    "    B, C_pooled, C, k_pooled = 1000, [], [], []\n",
    "    for b in 1:1:B\n",
    "        indices = rand(DiscreteUniform(1,size(dfprod,1)),Int64(round(size(dfprod,1)*0.8)))\n",
    "        dfprod_boot = dfprod[indices, :]\n",
    "        gamma11 = mean(dfprod_boot.e2)/mean(dfprod_boot.e11)\n",
    "        gamma21 = mean(dfprod_boot.e3)/mean(dfprod_boot.e21)\n",
    "        gamma31 = mean(dfprod_boot.e4)/mean(dfprod_boot.e31)\n",
    "        gamma32 = mean(dfprod_boot.e5)/mean(dfprod_boot.e32)\n",
    "        gamma22 = mean(dfprod_boot.e4)/mean(dfprod_boot.e22)\n",
    "        append!(G11,gamma11)\n",
    "        append!(G21,gamma21)\n",
    "        append!(G31,gamma31)\n",
    "        append!(G32,gamma32)\n",
    "        append!(G22,gamma22)\n",
    "        append!(mom1,1/(gamma11-1))\n",
    "        append!(mom2,2/(gamma21-1))\n",
    "        append!(mom3,3/(gamma31-1))\n",
    "        append!(mom4,3*gamma21/(gamma22*gamma11-gamma21))\n",
    "        append!(k_pooled,mean(dfprod_boot.e4)/(mean(dfprod_boot.e2))^2)\n",
    "    end\n",
    "    \n",
    "    # estimate the size of Gamma distribution using three different gammas\n",
    "    # weight myopically with inverse variance, standard errors from Delta method \n",
    "    Sigma31[1,1], Sigma31[1,2], Sigma31[1,3] = var(mom1), cov(mom1,mom2), cov(mom1,mom3)\n",
    "    Sigma31[2,1], Sigma31[2,2], Sigma31[2,3] = Sigma31[1,2], var(mom2), cov(mom2,mom3)\n",
    "    Sigma31[3,1], Sigma31[3,2], Sigma31[3,3] = Sigma31[1,3], Sigma31[2,3], var(mom3)\n",
    "    Sigma22[1,1], Sigma22[1,2], Sigma22[1,3] = var(mom1), cov(mom1,mom2), cov(mom1,mom4)\n",
    "    Sigma22[2,1], Sigma22[2,2], Sigma22[2,3] = Sigma22[1,2], var(mom2), cov(mom2,mom4)\n",
    "    Sigma22[3,1], Sigma22[3,2], Sigma22[3,3] = Sigma22[1,3], Sigma22[2,3], var(mom4)\n",
    "    \n",
    "    weights31, weights22 = ones(3)./[var(mom1), var(mom2), var(mom4)], ones(3)./[var(mom1), var(mom2), var(mom4)]\n",
    "    weights31, weights22 = weights31/sum(weights31), weights22/sum(weights22)\n",
    "\n",
    "    gamma11 = mean(dfprod.e2)/mean(dfprod.e11)\n",
    "    gamma21 = mean(dfprod.e3)/mean(dfprod.e21)\n",
    "    gamma31 = mean(dfprod.e4)/mean(dfprod.e31)\n",
    "    gamma32 = mean(dfprod.e5)/mean(dfprod.e32)\n",
    "    gamma22 = mean(dfprod.e4)/mean(dfprod.e22) # this is the underlying Kurtosis\n",
    "    M1, M2, M3, M4 = 1/(gamma11-1), 2/(gamma21-1), 3/(gamma31-1), 3*gamma21/(gamma22*gamma11-gamma21)\n",
    "    alpha_with31, alpha_with22 = [M1 M2 M3]*weights31, [M1 M2 M4]*weights22\n",
    "    alpha_with31, alpha_with22 = alpha_with31[1], alpha_with22[1]\n",
    "    \n",
    "    # write the rest of the statisticsistics\n",
    "    statistics.alpha_with31[i], statistics.alpha_with31_ste[i] = alpha_with31, sqrt((weights22')*Sigma31*weights22)\n",
    "    statistics.alpha_with22[i], statistics.alpha_with22_ste[i] = alpha_with22, sqrt((weights31')*Sigma22*weights31)\n",
    "    statistics.g11[i], statistics.g11_ste[i] = gamma11, std(G11)\n",
    "    statistics.g21[i], statistics.g21_ste[i] = gamma21, std(G21)\n",
    "    statistics.g22[i], statistics.g22_ste[i] = gamma22, std(G22)\n",
    "    statistics.g31[i], statistics.g31_ste[i] = gamma31, std(G31)\n",
    "    statistics.g32[i], statistics.g32_ste[i] = gamma32, std(G32)\n",
    "    statistics.M1[i], statistics.M2[i], statistics.M3[i], statistics.M4[i] = M1, M2, M3, M4\n",
    "    statistics.pooled_kurt[i], statistics.pooled_kurt_ste[i] = mean(dfprod.e4)/((mean(dfprod.e2))^2), std(k_pooled)\n",
    "    statistics.coefvar[i] = sqrt(mean(dfprod.e4)-(mean(dfprod.e2))^2)/mean((dfprod.e2))\n",
    "    statistics.implied_corr[i] = (statistics.pooled_kurt[i]/statistics.g22[i] - 1)/(statistics.coefvar[i]^2)\n",
    "    statistics.C[i] = statistics.C_pooled[i]*mean(num_correction)/(statistics.mean_inv_change[i]*statistics.std_price_change[i]^2)\n",
    "    \n",
    "    gammas[cat,1], gammas[cat,2:end] = i, [gamma11, gamma21, gamma22, gamma32]\n",
    "    cat += 1\n",
    "    print(\" category number \",(\" \"*string(c))[end-3:end],\" of the size \",length(ids),\"\\n\")\n",
    "end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "[2.09942, 12.1897, 228.677, 0.161035] 0.006755224324577662\n",
      "\n",
      "[1.05837, 6.02122, 91.4392, 0.109184] 0.002329063483591275\n",
      "\n",
      "[0.598555, 3.87261, 73.414, 2.51699e-8] 0.013485310293358999\n",
      "\n",
      "[1.84807, 9.77892, 173.048, 0.13125] 0.002279339438218855\n",
      "\n",
      "[3.12257, 9.8363, 0.627976, 0.580148] 8.984894451624026e-5\n",
      "\n",
      "[0.966629, 5.44156, 84.1539, 0.089239] 0.0010276647206831804\n",
      "\n",
      "[0.997788, 5.78315, 103.47, 0.0312284] 0.000574384098787137\n"
     ]
    }
   ],
   "source": [
    "# implement distance minimization for all categories\n",
    "# fit parameters of two Gamma distributions\n",
    "# alphas, the weight on the first one, and the ratio of betas are identified\n",
    "\n",
    "for cat in 1:1:7 \n",
    "    res = optimize(x->gap(x,gammas[cat,2:end]), [0.1, 0.1, 10.0, 0.0], autodiff=:forward)\n",
    "    i = Int64(gammas[cat,1])\n",
    "    print(\"\\n\",abs.(res.minimizer),\" \",res.minimum,\"\\n\")\n",
    "    statistics.alpha1[i] = abs(res.minimizer[1])\n",
    "    statistics.alpha2[i] = abs(res.minimizer[2])\n",
    "    statistics.ratio_betas[i] = abs(res.minimizer[3])\n",
    "    statistics.weight_on1[i] = abs(res.minimizer[4])\n",
    "end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10.777421896857614 0.10605110503016536 1.0000102251389038"
     ]
    }
   ],
   "source": [
    "# create an array of price changes for these categories\n",
    "# price changes in units of respective stadrard deviations\n",
    "\n",
    "standardized_changes = zeros(1)\n",
    "raw_changes = zeros(1)\n",
    "mean_duration = 0\n",
    "\n",
    "for c in statistics.category_id\n",
    "    dfcat = df2[df2.category .== c, :]\n",
    "    standardized_changes = vcat(standardized_changes,dfcat[dfcat.logchange.!=0,:logchange]/std(dfcat[dfcat.logchange.!=0,:logchange]))\n",
    "    raw_changes = vcat(raw_changes,dfcat[dfcat.logchange.!=0,:logchange])\n",
    "    mean_duration += sum(dfcat.duration)\n",
    "end\n",
    "\n",
    "mean_duration = mean_duration/length(raw_changes)\n",
    "N_pooled = 365/mean_duration\n",
    "V_raw = var(raw_changes[2:end])\n",
    "V_sta = var(standardized_changes[2:end])\n",
    "print(N_pooled,\" \",V_raw,\" \",V_sta)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the histogram and fitted density of non-standardized changes\n",
    "\n",
    "density = kde(raw_changes,bandwidth=0.035,kernel=Logistic)\n",
    "histogram(raw_changes,xlim=[-1.5,1.5],normed=true,bins=200,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the histogram and fitted symmetrized density of non-standardized changes\n",
    "\n",
    "raw_changes_sym = vcat(-raw_changes[2:end],raw_changes[2:end])\n",
    "density = kde(raw_changes_sym,bandwidth=0.035,kernel=Logistic)\n",
    "histogram(raw_changes,xlim=[-1.5,1.5],normed=true,bins=200,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 296,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the symmetrized histogram and fitted symmetrized density of non-standardized changes\n",
    "\n",
    "kernel_den_raw = DataFrame(x = density.x, q = density.density)\n",
    "CSV.write(\"kernel_den_raw.csv\",kernel_den_raw)\n",
    "histogram(raw_changes_sym,xlim=[-1.5,1.5],normed=true,bins=200,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 297,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the histogram and fitted density of standardized changes\n",
    "\n",
    "density = kde(standardized_changes,bandwidth=0.08,kernel=Logistic)\n",
    "kernel_den_raw = DataFrame(x = density.x, q = density.density)\n",
    "histogram(standardized_changes,xlim=[-4,4],normed=true,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 298,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the histogram and fitted symmetrized density of standardized changes\n",
    "\n",
    "standardized_changes_sym = vcat(-standardized_changes[2:end],standardized_changes[2:end])\n",
    "density = kde(standardized_changes_sym,bandwidth=0.08,kernel=Logistic)\n",
    "histogram(standardized_changes,xlim=[-4,4],normed=true,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "execution_count": 299,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the symmetrized histogram and fitted symmetrized density of standardized changes\n",
    "\n",
    "kernel_den_sta = DataFrame(x = density.x, q = density.density)\n",
    "CSV.write(\"kernel_den_sta.csv\",kernel_den_sta)\n",
    "histogram(standardized_changes_sym,xlim=[-4,4],normed=true,legend=false)\n",
    "plot!(density.x,density.density,color=:red,linewidth=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {},
   "outputs": [],
   "source": [
    "# produce the table with the statistics\n",
    "# part of this table is table 1 in the paper\n",
    "# another part goes to the apendix\n",
    "\n",
    "stat_string = DataFrame(category_id = Int64.(statistics.category_id))\n",
    "for name in names(statistics)\n",
    "    stat_string[!,name] = map(x->string(round(x;digits=3)),statistics[:,name])\n",
    "end\n",
    "stat_string[!, :number_price_changes] = Int64.(round.(statistics[!, :number_price_changes]))\n",
    "stat_string[!, :number_of_products] = Int64.(round.(statistics[!, :number_of_products]))\n",
    "stat_string[!, :category_id] = Int64.(round.(statistics[!, :category_id]))\n",
    "CSV.write(\"categories_big_table.csv\",stat_string);"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Julia 1.1.0",
   "language": "julia",
   "name": "julia-1.1"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "1.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
